A quantum transport barycentre is a quantum state chosen to minimize the weighted transport cost to a collection of other quantum states. It is not, in general, the matrix average of their density operators: the calculation optimizes over bipartite quantum couplings and cost operators. For Gaussian inputs with canonical quadratic costs, Gerolin and Lin show that the problem can be reduced to a finite-dimensional convex optimization over covariance matrices.
What is a quantum transport barycentre?
Let the input states be density operators σ1, …, σN, where σs acts on a Hilbert space Hs. Assign each input a nonnegative weight αs, with the weights summing to one. A candidate barycentre is a quantum state ρ on a common Hilbert space H0.
As an Amazon Associate I earn from qualifying purchases.
For each input, compare ρ with σs using a nonnegative self-adjoint cost operator Cs on the joint space H0 ⊗ Hs. The transport cost is the least expected cost over joint quantum states, or couplings, whose partial traces are ρ and σs. The barycentre minimizes the weighted sum of these individual transport costs.
Do these 3 things before closing this tab:
1Repair Windows errors before they cause bigger problems2Scan for outdated or missing drivers - takes under a minute3Clear out junk files and repair common Windows errorsIn compact form, the structure is:
- Choose a candidate state ρ on H0.
- For each input σs, minimize the expectation of Cs over joint states with marginals ρ and σs.
- Choose the ρ that minimizes the sum of those costs weighted by αs.
The exact answer depends on the chosen Hilbert spaces, cost operators, weights, and allowed class of candidate states. Those are part of the mathematical model, not incidental implementation details.
#1 Best Overall
Why it is not an average density matrix
The arithmetic mixture Σs αsσs is a valid quantum state, but that fact alone does not make it a transport barycentre. A transport barycentre is defined by minimizing a cost. The optimization accounts for how each candidate can be coupled to each input under the specified cost operators; it does not minimize entrywise distance between matrices.
The distinction is similar to the classical difference between an ordinary average and a Wasserstein barycentre. In classical optimal transport, a barycentre minimizes a weighted sum of transport costs, commonly written as Σs αsWpp(μ, νs). The underlying space, cost, exponent, and permitted barycentre class matter. In particular, a definition using powered distance Wpp should not be silently interchanged with one using the distance Wp.
How to calculate one
1. Specify the transport model
List the input density operators, their Hilbert spaces, their weights, the common barycentre space, and each cost operator. If using a 2-quantum Wasserstein formulation, state which canonical quadratic-cost convention applies. Gerolin and Lin’s framework covers both state-state and channel-based formulations, but the objects and constraints in those models are not interchangeable; the formulation must be identified before interpreting a result.
The Tool Desk
Outbyte Driver Updater FREEFix the driver behind crashes, sound loss and screen glitchesFind Drivers →Outbyte PC Repair FREEClear out junk files and repair common Windows errorsFree Scan →Rank #2
2. Set up the coupling constraints
For each input, consider joint states on H0 ⊗ Hs. Their partial trace over the input system must equal the candidate ρ, and their partial trace over the barycentre system must equal σs. Minimize the expected value of Cs among those admissible couplings. Then minimize the weighted sum of the resulting costs over candidate states ρ.
This is a nested variational problem over quantum states. Merely averaging input matrices skips the coupling constraints and the transport-cost minimization.
3. Check whether a minimizer is guaranteed to exist
Existence and duality results in Gerolin and Lin’s framework use hypotheses that include confinement and finite-cost feasibility. They should not be assumed automatically, especially for unbounded cost operators or continuous-variable systems. In those settings, verify that the states and costs meet the relevant assumptions before treating an infimum as an attained minimum or applying a dual formulation.
Rank #3
4. Use covariance optimization for the supported Gaussian case
When the inputs are Gaussian and the costs are canonical quadratic costs, Gerolin and Lin show that a Gaussian minimizer exists and that the minimum reduces to a finite-dimensional convex optimization over covariance matrices. This is the practical simplification: instead of optimizing over arbitrary quantum states and couplings, the Gaussian problem can be represented and solved at the covariance-matrix level under the paper’s assumptions.
The reduction is specific to that Gaussian and cost setting. It does not establish that the same covariance-only calculation solves the general problem for arbitrary input states or arbitrary cost operators.
5. Reconstruct the state and check uniqueness separately
A unique optimal covariance matrix does not by itself prove there is a unique quantum state realizing it. The authors use a state-reconstruction principle under covariance complementary slackness. They further establish that if at least one Gaussian input is faithful, the barycentre is unique among all quantum states and is necessarily Gaussian.
That is a sufficient condition stated in their result, not a claim that faithfulness is necessary for uniqueness in every instance. Without the required uniqueness argument, distinguish “the covariance optimizer is unique” from “the barycentre state is unique.”
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.What classical optimal transport contributes—and what it does not
Classical empirical Wasserstein barycentres provide useful intuition for the minimization structure, but their algorithms are not quantum barycentre solvers. For empirical measures, transport between supports can be expressed as a linear program over nonnegative coupling matrices with prescribed row and column marginals. Cuturi and Doucet describe classical methods that optimize barycentre weights with convex subgradient techniques when support is fixed, and alternating weight/location procedures for free support that can reach local minima.
Quick wins for a faster PC:
Fix the driver behind crashes, sound loss and screen glitchesFind Drivers →Repair Windows errors before they cause bigger problemsFix Now →Those procedures act on classical empirical measures and coupling matrices. They do not, by themselves, handle density operators, bipartite quantum states, or quantum cost operators.
When is a quantum Wasserstein barycentre unique?
Uniqueness depends on the formulation and hypotheses. For the Gaussian setting in their framework, Gerolin and Lin state that a faithful Gaussian input is sufficient to ensure a unique barycentre among all quantum states, with that barycentre necessarily Gaussian. Covariance-level uniqueness alone is a weaker conclusion unless a state-reconstruction argument connects the covariance optimizer to a unique state.
Gerolin and Lin’s work, “Quantum Optimal Transport Barycenters: Existence, Duality, and Gaussian Rigidity,” is an arXiv v1 preprint submitted October 1, 2026. Its existence, duality, Gaussian reduction, and uniqueness claims should therefore be attributed to the authors’ preprint rather than presented as settled textbook consensus.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
